Brain Injury Knowledge Ontology (BIKO) for traumatic brain injury: Formalizing concepts and methods used in translational traumatic brain injury research.

Traumatic brain injury (TBI) is an insult to the brain resulting from an external force and is a significant cause of morbidity and mortality in the United States. No effective clinical therapeutics currently exist for this injury. Although several therapies and procedures have been deemed successful for TBI treatment in preclinical research studies, they have yet to be translated into human patients. These discouraging results have left many scientists questioning the role of animal models in drug discovery for TBI. One major hurdle in translating the knowledge obtained in the laboratory to the clinic is the methodological variance across these studies. This variance can hinder the ability to draw conclusions from conflicting studies and aggregate data across various research studies, which ultimately impedes the ability to aggregate data across these studies. Therefore, addressing this variance is crucial for bridging the gap between the laboratory and the clinic. The increasing volume of papers and associated data being published every day makes this hurdle even more difficult to overcome. The initial steps to address these knowledge gaps are identifying these studies and creating a shared knowledge framework for mapping their terminology. We are developing the Brain Injury Knowledge Ontology (BIKO) to create a standardized model to describe methods and outcome measures used within preclinical and clinical TBI therapy studies to facilitate comparison across studies and models. The first version of BIKO focuses on modeling the major preclinical TBI models, e.g., Controlled Cortical Impact Model, Fluid Percussion Model, and Weight-Drop Model), major neurological injuries related to these models and their relationship to clinical pathophysiology. We show how BIKO provides a machine-readable way to represent the methodologies used in TBI therapeutic studies to compare models across clinically relevant features.